Fatigue and creep-fatigue deformation of several nickel-base superalloys at 650 °c

Specimens of seven nickel-base superalloys for gas turbine disk application that had been failed in fatigue and creep-fatigue at 650 °C were examined by transmission electron microscopy to observe the effects of composition and microstructure on the deformation characteristics of the alloys. The alloys were Waspaloy, HIP Astroloy, H+F Astroloy, H+F René 95, IN 100, MERL 76, and NASA IIB-7. The amount of bulk deformation observed in all the alloys was low. At inelastic strain amplitudes less than about 10^-3 only favorably oriented grains exhibited yielding, and the majority of those had 〈110〉 near the tensile axis. Deformation occurred on octahedral systems for all the alloys except MERL 76, which also exhibited primary cube slip. The difference in slip behavior between MERL 76 and its parent composition, IN 100, was attributed to the addition of Nb. Deformation occurred in well-defined slip bands in the alloys that contained only fine aging γ′, 0.01 to 0.06 μm in size. Alloys which also contained a population of larger aging γ′ particles, 0.1 to 0.3 μm, exhibited more homogeneous deformation. Deformation in the creep-fatigue cycle, which employed a 15 minute dwell at the maximum tensile strain of the cycle, was not greatly different from fatigue deformation except that a few extended faults were formed.     

Preparation and properties of fine grain β- CuAlNi strain- memory alloys

A method has been developed to produce grain sizes as small as 5 μm in alloys of β-CuAlNi. The alloys were of eutectoid composition and a procedure was developed for determining the composition of a eutectoid alloy having any required value for transition temperature (M _s ). The thermo-mechanical treatment involved two sequential stages of warm rolling followed by recrystallization. The alloys produced were single phase β-type with no second phase being present. Characteristic two-stage stress-strain curves were obtained for most of the specimens. It was generally found that the tensile strength and strain to failure increased with decreasing grain size according to a Hall-Petch type relationship down to a grain size of 5 μm. A fracture strength of 1200 MPa and a fracture strain of 10 pct were obtained in the best alloy. It was found that the major recovery mode, whether pseudoelastic or strain-memory, did not have any significant effect on the total recovery obtained. Recovery properties were not affected significantly by decreasing grain size, and 86 pct recovery could still be obtained at a grain size of around 10 μm. Grain refinement improved the fatigue life considerably, possibly due to the high ultimate fracture stress and ductile fracture mode. A fatigue life of 275,000 cycles could be obtained for an applied stress of 330 MPa and a steady state strain of 0.7 pct. At fine-grain sizes most of the fractures were due to transgranular-type brittle fracture and micro void-type ductile fracture, depending on the alloy composition. It was suggested that the difference between the alloys was due to differences in oxygen segregation at the grain boundaries.     

Hydrogen partitioning in pure cast aluminum as determined by dynamic evolution rate measurements

A statistical analysis of the porosity in 99.995 wt pct pure commercially available cast aluminum has been correlated with real time hydrogen evolution data obtained in an ultrahigh vacuum furnace in order to estimate the hydrogen partitioning in the aluminum. The dynamic technique employed permitted the detection and separation of hydrogen evolved from solid solution, hydrogen released by the rupture of large pores, and gases desorbed from the aluminum surface. Results of the statistical analysis indicate average pore diameters in pure cast aluminum extend from less than 1 to over 400 μm. Interdendritic pores having diameters greater than 25 μm constitute over 98 pct of the pore volume. The overall volume fraction of pores was determined to be 0.71 pct. Compared to vacuum remelted rolled aluminum, the porosity resulted in a reduction of ultimate tensile strength of 13 pct and a reduction in yield strength of 21 pet. The evolution of hydrogen from the aluminum was observed to occur by large hydrogen pressure pulses due to the rupture of pores near the surface and by a smooth steady desorption from solid solution. The rupturing pores were observed visually and found to occur both in the solid state and after melting. A substantial change in slope of the desorption curve following the pulse train suggests the pores are the primary sources of hydrogen in the bulk. Analysis of the pore and pulse size distributions indicates more than 99 pct of the bulk hydrogen is partitioned in pores greater than 25 μm. Pressures within the larger pores (≈270 μm) were determined to be about 2.4 atm at room temperature. Hydrogen content in the large pores was found to be as high as 2 × 10¹⁶ molecules. The total hydrogen content in the pores and in solid solution was determined to be 6.3 × 10¹⁷ atoms/cm³ (0.43 cm³/100 g). Measurements on commercially available 99.9995 wt pct cast aluminum indicate the total hydrogen content to be 4.8 × 10¹⁷ atoms/cm³ (0.33 cm³/100 g).     

Superplastic behavior of two-phase titanium aluminides

A two-phase Ti(57 at. pct)-Al(43 at. pct) alloy with an initial lamellar microstructure was thermomechanically processed to form an equiaxed fine-grained structure. The fine-grained (- L = 5 μm) material was superplastic in the temperature range 1000 °C to 1100 °C, exhibiting a stress exponent of about 2 with a tensile ductility of 275 pct. The rate-controlling deformation mechanism is proposed to be grain boundary sliding accommodated by slip controlled by lattice diffusion in TiAl. At room temperature, the lamellar and fine-grained materials exhibit the same compressive yield stress. The compressive strain to failure, however, for the fine-grained material was about 28 pct compared to 6 pct for the lamellar material.     

Structure and properties of a splat cooled 2024 aluminum alloy

Pound lots of splat cooled 2024 aluminum flake materials were produced by rapidly quenching the atomized melt against a rotating copper disc. Three flake sizes were selected, cold compacted into aluminum cans, and extruded at 300°C at a reduction ratio of 20 to 1. The extruded rods were reduced 50 pct by cold swaging, solution treated at 495°C, water quenched, and naturally aged. The splat cooled 2024 alloy had constituent particles of 1 fim and finer (compared to 5 to 20 μm for the commercial alloy); further, one of the complex constituent phases (AlCuFeMn) was essentially eliminated by the rapid quench. Compared to commercial 2024-T4, the splat cooled 2024 alloys showed 14 to 17 pct increase in yield and tensile strength (no loss of ductility) a seven-fold increase in fatigue life at 30,000 psi, and a large improvement in the 300°F (150°C) stress rupture life in tests beyond 100 h. The fracture characteristics of the splat alloys, while exhibiting excellent to superior ductility, appear inhomogeneous due to the presence of finely dispersed oxide films scattered in the structure. Formerly Research Assistant, Department of Metallurgy and Materials Science, Massachusetts Institute of Technology     

Corrosion fatigue in nitrocarburized quenched and tempered steels

In order to investigate the fatigue strength and fracture mechanism of salt bath nitrocarburized steels, specimens of the steels SAE 4135 and SAE 4140, in a quenched and tempered state, and additionally in a salt bath nitrocarburized and oxidizing cooled state as well as in a polished (after the oxidizing cooling) and renewed oxidized state, were subjected to comparative rotating bending fatigue tests in inert oil and 5 pct NaCl solution. In addition, some of the quenched and tempered specimens of SAE 4135 material were provided with an approximately 50-μm-thick electroless Ni-P layer, in order to compare corrosion fatigue behavior between the Ni-P layer and the nitride layers. Long-life corrosion fatigue tests of SAE 4135 material were carried out under small stresses in the long-life range up to 10⁸ cycles with a test frequency of 100 Hz. Fatigue tests of SAE 4140 material were carried out in the range of finite life (low-cycle range) with a test frequency of 13 Hz. The results show that the 5 pct NaCl environment drastically reduced fatigue life, but nitrocarburizing plus oxidation treatment was found to improve the corrosion fatigue life over that of untreated and Ni-P coated specimens. The beneficial effect of nitrocarburizing followed by oxidation treatment on cor-rosion fatigue life results from the protection rendered by the compound layer by means of a well-sealed oxide layer, whereby the pores present in the compound layer fill up with oxides. The role of inclusions in initiating fatigue cracks was investigated. It was found that under corrosion fatigue conditions, the fatigue cracks started at cavities along the interfaces of MnS inclusions and matrix in the case of quenched and tempered specimens. The nitrocarburized specimens, however, showed a superposition of pitting corrosion and corrosion fatigue in which pores and nonmetallic inclusions in the compound layer play a predominant role concerning the formation of pits in the substrate.     

Effects of composition and microstructure on fatigue of γ/γ′-δ type eutectic alloys

Room temperature tension-tension fatigue tests were performed on two lamellar γ/γ′-δ alloys, one with 0 pct Cr and one with 6 pct Cr. The 6 pct Cr alloy was solidified at 3 cmJh while the 0 pct Cr alloy was solidified at 3 cm/h and 5.7 cm/h. Fatigue testing was done on both alloys in the as-directionally solidified condition and on the 0 pct Cr alloy after heat treatment. Increasing the growth speed of the 0 pct Cr alloy increased the fatigue life of the material at stresses above the 10⁷ cycle fatigue limit. Partial solution treating and aging of the 0 pct Cr alloy,R = 3 cm/h, increased the fatigue life relative to the as-directionally solidified material at high stresses, to the same extent as increasing the growth speed. Full solution treatment and aging of the 0 pct Cr alloy,R = 5.7 cm/ h, caused a reduction in the fatigue life relative to the as-directionally solidified material. Fatigue cracking tended to be faceted in the 6 pct Cr alloy as opposed to the more ductile failure of the 0 pct Cr alloy. Microstructural perfection, grain size and shape, interlamellar spacing, longitudinal cracking, and longitudinal and transverse ductility all are believed to have influenced the fatigue resistance of the alloys.     

Measurement of fatigue accumulation in high-strength steels by microstructural examination

Fatigue test bars fabricated from an SA508 class 3 low-carbon steel plate were cyclically deformed at 300 °C (constant low-cycle fatigue, total strain range Δε = 0.78 pct and 0.48 pct) to crack initiation (100 pct cumulative damage, CD) and to the factors 75, 50, and 25 pct CD. The test bars were cut perpendicular to the stress axis at the center of the gage length. The X-ray diffraction line-broadening (XRD) was performed on the cross sections created by the cuts. Thin foils (∼0.1-μm thick) were prepared from each cross section and used for the transmission electron microscope (TEM) and selected area diffraction (SAD) study. The half-value line breadth change measured by the XRD increased with the CD increase up to 50 pct, beyond which a significant reduction was observed for the 75 and 100 pct CD sample regardless of the incident X-ray beam angle. By the TEM, the undamaged material (0 pct CD) was characterized by high-angle boundaries, small carbide precipitates, and dislocation cell networks in grains. These characteristics did not show any appreciable changes in all of the samples with fatigue damage of the respective levels. Micro-orientation changes of the dislocation cells studied by the SAD of the foils and a statistical data analysis clearly demonstrated that the mean orientation difference in the cells and its standard deviation increased gradually as the CD increased.     

Isothermal fatigue of low tin lead based solder

Low tin lead based solder fails by intergranular and/or transgranular modes depending upon experimental conditions. At low frequency and in tests with hold times separation of grains is the main mode of fracture. In the 5 to 100 °C temperature range at high frequency (> 10⁻² Hz) and at high total strain range (0.75 pct) the failure mode is mixed transgranular-intergranular; at a low total strain range (0.3 pct) the mode of failure is intergranular. Change in failure mode leads to a bend in the Coffin-Manson plot. Tensile hold time and combined tensile and compressive hold times are found to reduce dramatically the fatigue cycles to failure of this solder. A simple mathematical relation between the fatigue life of the solder and ramp time, tensile, and compressive hold times is developed.     

Identification of rolling-sliding damage mechanisms in porous alloys

Lubricated rolling-sliding damage in a relatively soft Fe-0.6 pct C alloy and a relatively hard carbonitrided iron, both produced by powder metallurgy, has been investigated. Damage mechanisms were controlled by large-scale as well as small-scale plastic deformations. A large-scale, bulk plastic deformation process produced surface densification in the Fe-0.6 pct C alloy. Formation of surface cracks by asperity-scale plastic shearing was also observed in both materials. Small-scale plastic deformation processes, restricted to the pore edges, gave rise to the formation of fatigue microcracks at the boundary between the densified and undensified region in the Fe-0.6 pct C alloy. A similar effect was found at a depth of between 550 and 1000 μm in the carbonitrided material. Moreover, in the Fe-0.6 pct C alloy, these plastic deformations also triggered the formation and propagation of macrocracks, which produced macroscopic damage by spalling. The damage mechanisms due to small-scale plastic deformations were explained on the basis of a local approach model, able to account for the influence of pores on the mechanical behavior of the materials. However, this approach could not explain the microcracks, which were found at the surface pores in the carbonitrided material. Their formation was ascribed to the interplay between the surface tensile (friction) stresses and the low matrix toughness of the material near the surface.     

The effect of environment on high-temperature hold time fatigue behavior of annealed 2.25 pct Cr 1 pct Mo steel

Total strain control fatigue tests with a 120-second hold period at either peak compressive or tensile strain were conducted on annealed 2.25 pct Cr 1 pct Mo steel. Tests were performed at the total strain range of 1.0 pct at 500 °C or 600 °C in air, 1.3 Pa (10⁻² torr) or 1.3 × 10⁻³ Pa (10⁻⁵ torr) vacuum. The nature of the hold and the environment affect fatigue life and surface crack patterns. A compressive hold is more deleterious than a tensile hold in high-temperature air, while the reverse is true in environments in which oxidation is limited. Observations of cracks at the surface and in cross section indicate that an oxidation-fatigue interaction accounts for the damaging effect of a compressive hold in air tests. In vacuum tests, creep damage has the opportunity to accumulate and causes the tension hold to exhibit the shortest fatigue lifetime.     

Isothermal fatigue of an aluminide-coated single-crystal superalloy: Part I

The isothermal fatigue behavior of a high-activity aluminide-coated single-crystal superalloy was studied in air at test temperatures of 600 °, 800 °, and 1000 °. Tests were performed using cylindrical specimens under strain control at ≈0.25 Hz; total strain ranges from 0.5 to 1.6 pct were investigated. At 600 °, crack initiation occurred at brittle coating cracks, which led to a significant reduction in fatigue life compared to the uncoated alloy. Fatigue cracks grew from the brittle coating cracks initially in a stage II manner with a subsequent transition to crystallographic stage I fatigue. At 800 ° and 1000 °, the coating failed quickly by a fatigue process due to the drastic reduction in strength above 750 °, the ductile-brittle transition temperature. These cracks were arrested or slowed by oxidation at the coating-substrate interface and only led to a detriment in life relative to the uncoated material for total strain ranges of 1.2 pct and above 800 °. The presence of the coating was beneficial at 800 ° for total strain ranges less than 1.2 pct. No effect of the coating was observed at 1000 °. Crack growth in the substrate at 800 ° was similar to 600 °; at 1000 °, greater plasticity and oxidation were observed and cracks grew exclusively in a stage II manner. Formerly Research Student, Department of Materials Science and Metallurgy, University of Cambridge. Formerly Lecturer, Department of Materials Science and Metallurgy, University of Cambridge CB2 3QZ, United Kingdom.     

Electron beam surface modification of a porous bronze-graphite composite

Elemental powders were mixed to obtain a 90 wt pct copper, 8 wt pct tin, and 2 wt pct graphite composite. The porosity level of the sintered specimens was reduced from 25 to 10 pct, which resulted in an increase in the macrohardness value from 17 Hv (90 MPa) to 67 Hv (355 MPa); the density of the sintered specimen was 7.80 g · cm⁻³. The synthesized material was then subjected to electron beam (EB) surface melting. The resultant surface was homogeneous and the microstructural features were refined. The segregation level and variation in the microhardness were drastically reduced. The morphology of the otherwise irregular pores changed to spherical, thereby reducing their interfacial energy. An intriguing modification in the EB melted layer had a density gradient with depth that is sensitive to the heating time of the material using EB. At a heating time of 250 ms, the upper region of the melted layer was dense and hard; the density and the hardness were 8.5 g · cm⁻³ and 103 ± 7 Hv, respectively, while the lower region had density of 6.7 g · cm⁻³ (porosity 22 pct). If the heating time was reduced to 17 ms, the distribution of pores was reversed; the density of upper and lower layers changed to 3.9 and 8.2 g · cm⁻³, respectively. In spite of the higher density of pores, the EB processed composite exhibited increased hardness, compressive strength, and tensile strength. The formation of pores in the lower EB melted region was explained using a qualitative fluid flow model. The combination of a dense substrate and porous surface was desirable, since the former improved the strength and the thermal conductivity of the composite and the latter could be impregnated with oil to achieve the required lubrication levels.     

Effect of strain rate on the high-temperature low-cycle fatigue properties of a nimonic PE-16 superalloy

Strain-rate effects on the low-cycle fatigue (LCF) behavior of a NIMONIC PE-16 superalloy have been evaluated in the temperature range of 523 to 923 K. Total-strain-controlled fatigue tests were performed at a strain amplitude of ±0.6 pct on samples possessing two different prior microstructures: microstructure A, in the solution-annealed condition (free of γ′ and carbides); and microstructure B, in a double-aged condition with γ′ of 18-nm diameter and M₂₃C₆ carbides. The cyclic stress response behavior of the alloy was found to depend on the prior microstructure, testing temperature, and strain rate. A softening regime was found to be associated with shearing of ordered γ′ that were either formed during testing or present in the prior microstructure. Various manifestations of dynamic strain aging (DSA) included negative strain rate-stress response, serrations on the stress-strain hysteresis loops, and increased work-hardening rate. The calculated activation energy matched well with that for self-diffusion of Al and Ti in the matrix. Fatigue life increased with an increase in strain rate from 3 × 10^-5 to 3 × 10^-3 s^-1, but decreased with further increases in strain rate. At 723 and 823 K and low strain rates, DSA influenced the deformation and fracture behavior of the alloy. Dynamic strain aging increased the strain localization in planar slip bands, and impingement of these bands caused internal grain-boundary cracks and reduced fatigue life. However, at 923 K and low strain rates, fatigue crack initiation and propagation were accelerated by high-temperature oxidation, and the reduced fatigue life was attributed to oxidation-fatigue interaction. Fatigue life was maximum at the intermediate strain rates, where strain localization was lower. Strain localization as a function of strain rate and temperature was quantified by optical and scanning electron microscopy and correlated with fatigue life.     

Comparison of the Microstructure, Tensile, and Creep Behavior for Ti-22Al-26Nb (At. Pct) and Ti-22Al-26Nb-5B (At. Pct)

The effect of the addition of 5 at. pct boron on the microstructure and creep behavior of a nominally Ti-22Al-26Nb (at. pct) alloy was investigated. The boron-modified alloy contained boride needles enriched in titanium and niobium, and because to these borides, this material was considered to be a discontinuously reinforced metal matrix composite. These needle-shaped borides made up to 2 pct of the volume and were up to 158-μm long and 22-μm wide. The effect of boron on the mechanical properties was evaluated through in-situ creep testing and tensile testing at room temperature (RT) and 650 °C. Overall, the addition of 5 at. pct boron proved to be detrimental to the tensile and creep behavior. The composite exhibited a brittle failure and lower elongations-to-failure than the monolithic material. The in-situ tensile and creep experiments revealed that the deformation process initiated in the boride needles, which cracked extensively, and significantly greater primary creep strains and creep rates were exhibited by the composite.

---

Mechanical properties and fracture behavior of an ultrafine-grained Al-20 wt pct Si alloy

The effect of powder particle size on the microstructure, mechanical properties, and fracture behavior of Al-20 wt pct Si alloy powders was studied in both the gas-atomized and extruded conditions. The microstructure of the as-atomized powders consisted of fine Si particles and that of the extruded bars showed a homogeneous distribution of fine eutectic Si and primary Si particles embedded in the Al matrix. The grain size of fcc-Al varied from 150 to 600 nm and the size of the eutectic Si and primary Si was about 100 to 200 nm in the extruded bars. The room-temperature tensile strength of the alloy with a powder size <26 μm was 322 MPa, while for the coarser powder (45 to 106 μm), it was 230 MPa. The tensile strength of the extruded bar from the fine powder (<26 μm) was also higher than that of the Al-20 wt pct Si-3 wt pet Fe (powder size: 60 to 120 μm) alloys. With decreasing powder size from 45 to 106 μm to <26 μm, the specific wear of all the alloys decreased significantly at all sliding speeds due to the higher strength achieved by ultrafine-grained constituent phases. The thickness of the deformed layer of the alloy from the coarse powder (10 μm at 3.5 m/s) was larger on the worm surface in comparison to the bars from the fine powders (5 μm at 3.5 m/s), attributed to the lower strength of the bars with coarse powders.     

An assessment of cold work effects on strain-controlled low-cycle fatigue behavior of type 304 stainless steel

The influence of prior cold work (PCW) on low-cycle fatigue (LCF) behavior of type 304 stainless steel has been studied at 300, 823, 923, and 1023 K by conducting total axial strain-controlled tests in solution annealed (SA, 0 pct PCW) condition and on specimens having three levels of PCW, namely, 10, 20, and 30 pct. A triangular waveform with a constant frequency of 0.1 Hz was employed for all of the tests performed over strain amplitudes in the range of ±0.25 to ± 1.25 pct. These studies have revealed that fatigue life is strongly dependent on PCW, temperature, and strain amplitude employed in testing. The SA material generally displayed better endurance in terms of total and plastic strain amplitudes than the material in 10, 20, and 30 pct PCW conditions at all of the temperatures. However, at 300 K at very low strain amplitudes, PCW material exhibited better total strain fatigue resistance. At 823 K, LCF life decreased with increasing PCW, whereas at 923 K, 10 pct PCW displayed the lowest life. An improvement in life occurred for prior deformations exceeding 10 pct at all strain amplitudes at 923 K. Fatigue life showed a noticeable decrease with increasing temperature up to 1023 K in PCW state. On the other hand, SA material displayed a minimum in fatigue life at 923 K. The fatigue life results of SA as well as all of the PCW conditions obeyed the Basquin and Coffin-Manson relationships at 300, 823, and 923 K. The constants and exponents in these equations were found to depend on the test temperature and prior metallurgical state of the material. A study is made of cyclic stress-strain behavior in SA and PCW states and the relationship between the cyclic strain-hardening exponent and fatigue behavior at different temperatures has been explored. The influence of environment on fatigue crack initiation and propagation behavior has been examined.     

Low-Cycle Fatigue Behavior of Die-Cast Mg Alloys AZ91 and AM60

The influence of microstructure and artificial aging response (T6) on the low-cycle fatigue behavior of super vacuum die-cast (SVDC) AZ91 and AM60 has been investigated. Fatigue lifetimes were determined from the total strain-controlled fatigue tests for strain amplitudes of 0.2?pct, 0.4?pct, 0.6?pct, 0.8?pct, and 1.0?pct under fully reversed loading at a frequency of 5?Hz. Cyclic stress?Cstrain behavior was determined using an incremental step test (IST) and compared with the more traditional constant amplitude test. Two locations in a prototype casting were investigated to examine the role of microstructure and porosity on fatigue behavior. At all total strain amplitudes microstructure refinement had a negligible impact on fatigue life because of significant levels of porosity. AM60 showed an improvement in fatigue life at higher strain amplitudes when compared with AZ91 because of higher ductility. T6 heat treatment had no impact on fatigue life. Cyclic stress?Cstrain behavior obtained via the incremental step test varied from constant amplitude test results due to load history effects. The constant amplitude test is believed to be the more accurate test method. In general, larger initiation pores led to shorter fatigue life. The fatigue life of AZ91 was more sensitive to initiation pore size and pore location than AM60?at the lowest tested strain amplitude of 0.2?pct. Fatigue crack paths did not favor any specific phase, interdentritic structure or eutectic structure. A multistage fatigue (MSF) model showed good correlation to the experimental strain-life results. The MSF model reinforced the dominant role of inclusion (pore) size on the scatter in fatigue life.     

The Study of Fatigue Behaviors and Dislocation Structures in Interstitial-Free Steel

There are three types of cyclic hardening for cyclically deformed interstitial-free (IF) steels. The magnitude of cyclic hardening was unobvious and dislocation cells smaller than 2 μm were very hard to find when total strain amplitude (Δε/2) was controlled to within 0.1 pct. When Δε/2 is increased to 0.125 to 0.3 pct, secondary cyclic hardening takes place prior to fatigue failure. Δε/2 = 0.6 pct, following an initial rapid-hardening stage. Dislocation cells smaller than 2 μm tend to develop near grain boundaries and triple junction of the grains while cycling just above Δε/2 = 0.125 pct. Such dislocation development results in secondary hardening. However, no failure occurs if cycling just below Δε/2 = 0.1 pct; hence, the fatigue limit for IF steel should be very close to Δε/2 = 0.1 pct.